On Fri, 10 Sep 2004 08:45:59 -0600, "Jeff Crowell"
wrote:

phil hunt wrote:

Simulators -- assuming a good mathematical model of the airplane --
should be able to correctly simulate how it would respond to
anything the pilot does. The visual part of simulation is mostly
solved these days due to good computer power. The hard thing, as I
see it, is simulating the effect of the aircraft's movements on the
pilot.

A very nontrivial challenge.

When positive G is modeled by inflating your g-suit and negative G
by inflating a "whoopie cushion" under the driver's butt or dropping
the sim a foot or two, that ain't very useful. Numerous crashes
have been attributed to pilots flying the airplane too soon after being
in the sim (Miramar had a mandatory delay between 'flying' the
WST and getting in a real airplane). Your body gets used to what
ought to happen to it in the Real Thing (tm), then gets confused by
the sim. Minutia such as rate of G application get missed by the sim
but have tremendous significance in flight.

Sims are great for buttonology and procedures, and can be a lot of
fun (and they can scare the hell out of you sometimes). But they do
NOT teach you how to really push the plane to its and your limits
(low-level flight in a non-permissive environment, for one simple
example), and that's the key to surviving in the Real World.

We've seen it again and again--try to save money in the training
environment and you guarantee increased losses in combat.


I agree to a point. It's a difficult task to simulate accelerations on
the body that occur in flight using some sort of six-degree of motion
ground-based gadget. It works fairly well in low acceleration systems
such as air transports, but not in high-g operations like tactical
aircraft.

But (you were waiting for that, I know), a lot depends upon what you
are trying to train. One can do a pretty good job of cockpit
procedures training without much high-tech whiz-bang. And, one can
teach instrument procedures pretty well with moderate tech sims. And,
if you spend the money, current state-of-the-art can give you a pretty
good aircraft pilot qual without ever burning a pound of JP-8.

It's when you get into the weapons employment phase that things get
confusing. Exactly as you describe, there's the proprioceptive cues
that are part and parcel of every highly qualified operators input.
You can't recreate those (yet) with the desired level of accuracy.
And, you can't--without huge investment--recreate the total combat
environment. You can't get the total combination of airplane, flight,
strike package, support systems, enemy counter, enemy sensors, enemy
IADS, electronics, etc. etc. etc. For that matter, you can't very
easily or economically do "war" in training.

One of the things we were working on with the ATF (F-23) program was
low-cost desk-top trainers networked with both dome simulators and
computer-generated entities to create a combat scenario. While the
fidelity was unbelievably low if compared to actual flight, the task
wasn't to teach airplane/weapon operation but to try to teach
situational awareness--that "big-picture" or sixth sense that good air
warriors carry in their heads.

Surprisingly, a group of Fighter Weapons School, Top Gun, flight test
and operational USAF/USN aviators quickly found that they could get
immersed in the battle and almost forget that they were sitting at a
25" video monitor with a stick grip mounted on a desktop.

I used to compare it to watching a football game on a small screen TV.
Once you start watching you will often forget how small the display is
and you're simply concentrating on the game.

Tactics, maneuver, weapons employment, flight management, navigation,
systems operations, etc. could all be practiced. The only thing that
was missing was basic "stick-and-rudder".



Ed Rasimus
Fighter Pilot (USAF-Ret)
"When Thunder Rolled"
"Phantom Flights, Bangkok Nights"
Both from Smithsonian Books
***www.thunderchief.org